Jump to ContentJump to Main Navigation
Principles, Procedure, and JusticeEssays in honour of Adrian Zuckerman$

Rabeea Assy and Andrew Higgins

Print publication date: 2020

Print ISBN-13: 9780198850410

Published to Oxford Scholarship Online: June 2020

DOI: 10.1093/oso/9780198850410.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (oxford.universitypressscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use. date: 07 August 2020

How Judges Decide

How Judges Decide

(p.91) 5 How Judges Decide
Principles, Procedure, and Justice

Robert J Sharpe

Oxford University Press

Abstract and Keywords

This chapter considers what HLA Hart described as a choice between the two extremes of the realist ‘nightmare’ that judges never decide according to the law and the idealist ‘noble dream’ that judges always decide according to the letter of the law. The chapter explores the reasons for legal uncertainty and examines the constraints that control judicial decision-making. Law is uncertain because it is necessarily general and its application in any particular case depends upon the context. This means that judges often have a choice but that choice is constrained by several factors, especially by the obligation to provide a reasoned judgment.

Keywords:   judicial decision-making, legal uncertainty, judicial choice, disciplined decision-making, reasons for judgment


In this chapter, I discuss some of factors that constrain and control judicial decision-making. The law is not always clear and while judges are sworn to follow the law, the law often does not provide them with clear guidance. I think that it is important for judges to admit that they have to make choices, to confront what influences them to decide this way or that, and to advert to the constraints on their decision-making authority. My topic gives rise to theoretic issues about the nature of law and the role of judges, but I write from the perspective of a sitting judge, not a legal philosopher. As seems appropriate for a volume dedicated to a great scholar of procedural justice, my focus will be on procedural values that constrain judges when they make decisions, the most important of which is the right of every litigant to a reasoned decision.

I begin by discussing the phenomenon of legal uncertainty and the challenge that feature of the law poses for judges, particularly in an era when most judges have lost faith in the narrowly legalistic version of formalism that formerly prevailed. I turn to the problem posed by ‘hard cases’ for which the law does not appear to provide a clear answer and consider the approach a judge should take to such a case. After examining formal constraints, such as the obligation to base decisions on accepted legal sources, I turn to procedural values that limit the capacity of judges to decide cases on the basis of their own personal views. The adversarial values of party definition and party presentation significantly constrain judicial choice. The enforcement of the boundaries of decision-making authority, limiting the capacity of appellate judges to interfere with the factual and discretionary decisions of trial judges, represents a significant constraint on the choices appellate judges make, while the control appellate judges exercise over the law constrains trial judges. Finally, and most importantly, I turn to the right of every litigant to a reasoned decision. I argue that the obligation to give reasons and thereby (p.92) expose judicial thinking to public scrutiny represents the most important constraint on judicial choice and one that operates to ensure the integrity of judicial decision-making.

Legal uncertainty

The very idea of law suggests the need for clear rules capable of producing just and predictable results to govern society and social relations. Yet one of the greatest judges of the twentieth century, Benjamin Cardozo, wrote: ‘I was much troubled in spirit, in my first years on the bench, to find how trackless was the ocean on which I had embarked. I sought for certainty. I was oppressed and disheartened when I found the quest for it was futile.’1 The ideal system, said Cardozo, would be ‘a code at once so flexible and so minute, as to supply in advance for every conceivable situation the just and fitting rule’.2 But that level of perfection, Cardozo realized, is unattainable. The judge is often confronted with the task of deciding cases for which the law seems to provide no clear answer. A different but related problem arises where the law does provide a clear answer but the result appears to the judge to be unjust.

The twin phenomena of legal uncertainty and laws that sometimes produce unjust results arise from what I believe to be an essential characteristic of law itself. Laws must be framed in terms of norms or standards that have general application. Aristotle observed that ‘all law is universal’.3 Generality gives the law its objective, rational, and systematic quality. It is what distinguishes the law, on the one hand, from judicial decisions in specific cases, on the other.

Laws set standards but they do not decide specific cases. To determine whether or how the law applies to a particular case requires the exercise of judgment. Someone must decide what the facts are, identify the relevant legal norm or standard, and then determine whether and to what extent the legal norm or standard governs. That is the role of the judge: to interpret the standard and decide whether or not it applies to the facts of the case. The need for judicial fact-finding and judicial interpretation injects a significant element of uncertainty into the law.

Aristotle also recognized that although universality is an identifying feature of law, it can also be the source of injustice. ‘The law’, he wrote, ‘takes account of the majority of cases, although not unaware that in this way errors are made’. Despite the errors, ‘the law is none the less right; because the error lies not in the law nor in the legislator, but in the nature of the case; for the raw material of human behaviour (p.93) is essentially of this kind’.4 In this passage, Aristotle identified a central dilemma in achieving justice under law. Universality of application is a necessary attribute of law and ensures that most cases will be justly decided. But because of the variability of the ‘raw material of human behaviour’, the universal legal norm will not yield just results in all cases. Needless to say, judges have a strong inclination to decide cases justly. One of the central conundrums of judging is knowing when it is best to accept a result that seems unjust in order to uphold the integrity of the universal standard, and when the standard leaves sufficient room for the judge to avoid the apparent injustice.

Law in context and legal formalism

Law is uncertain because its application is contextual. As a judge, I do not see the law as an abstraction but rather as a tool I work with to decide disputes in a practical way that achieves social peace and harmony in the real world. The proper interpretation and application of legal rules depend upon the context. To take a familiar example, the precise meaning of the law’s promise of equality changes as social conditions and attitudes evolve. Same-sex relationships, not long ago criminalized, now enjoy legal protection. How did that happen? Equality has long been a cornerstone of the rule of law but its meaning takes on a changing hue as social attitudes and practices evolve.

But if the answer to the question ‘what does the law say about this?’ is always ‘that depends upon the context’, is the law rendered unstable and uncertain, and is there too much that is left subject to the perceptions, beliefs, biases, and prejudices of the judge deciding the case?

The narrow legalistic version of formalism that prevailed well into the twentieth century and that, to some extent, survives to this day, resists the idea that the meaning of the law is shaped by context. Legal formalism conceives the law to be an objective and internally coherent body of rules and principles that can be objectively derived from the text of statutes and precedents and applied deductively to decide disputes without reference to the social or political context.

While I accept that legal reasoning has a formal structure to which judges must adhere, my experience of hearing and deciding cases tells me that the law does not operate in a vacuum and that rigid adherence to the narrow legalistic formalist model misses or, even worse, conceals things that are very real and present in my work as a judge. The legal disputes I decide arise from the realities of daily life and from a wide range of changing social, economic, and political contexts. The legal texts and doctrines that I am sworn to follow and to apply, be they common law (p.94) or statutory in origin, are imbued with moral, philosophical, and social values. To pretend that the law can be mechanically applied in a purely objective, morally neutral manner would be to ignore the important question of what values actually drive and determine decisions. Judges do not decide cases entirely based on neutral, objective principles and to pretend that they do so conceals a significant component of judicial reasoning and discourages judges from reflecting upon and questioning the values they do apply.

The attack on legal formalism

For the past century, there have been waves of legal thinkers who have challenged the narrow legalistic version of the formalist ideal. The American legal realist movement of the 1920s to the 1940s undermined confidence in formalism. Realists argued that judicial decision-making essentially turns on the facts and on the personal or political biases of judges. The realists adopted an instrumentalist view that regarded law as a tool to serve broader social purposes, perhaps reflecting American legal culture’s tendency to see litigation as a means of achieving legal and social reform. The realists urged greater reliance on the use of social science insights to enhance the quality of judicial decisions and to advance social, economic, and political goals.5

The legal realists were the forerunners of critical legal studies, critical race theory, and feminist legal theory. These diverse schools of thought are premised on progressive political attitudes and claim that conventional formal legal reasoning is incoherent, indeterminate, and based upon raw political choice. The critical legal studies movement took flight in the 1970s. Critical legal studies scholars vigorously rejected the formalist vision of law and asserted that law is radically indeterminate. They argued that as rights can be manipulated, the legal process is not a neutral tool to resolve differences on a principled basis but rather a political device to ensure the protection of vested interests.6 Critical feminist and critical race theorists borrow, to some extent, the rhetoric of critical legal studies, but focus specifically on the evils of sexism and racism, and on the need to transform social and legal attitudes in order to achieve genuine equality.7

(p.95) From the other end of the political spectrum emerged the law and economics movement which shares the realist perception of law as lacking coherence or credibility as an autonomous discipline. Law and economics proponents view legal disputes as being nothing more than a point at which disparate forces and ideas compete for acceptance. Building on the utilitarian philosophy of Bentham and Mill, law and economics scholars argue that legal rules could best be explained, understood, and developed in terms of their attainment of economic efficiency.8

While these schools of thought have achieved less traction in Canada and the United Kingdom than in the United States, I think it would be a serious error to discount their influence. Many aspects of the once-radical realist attack have been more or less absorbed into conventional legal thinking. Most lawyers and judges now accept the notion that the law is imbued with moral, political, and social values. We accept as well that law and legal rules should be viewed and shaped in terms of the broader social, political, and economic context in which they operate. We no longer see law as a self-contained end in itself but rather as a tool that can be used to achieve social harmony, peace, and justice. We recognize that this renders the law less than certain and that there are hard cases in which the legal outcome is difficult to predict.

But if the law is uncertain in its application and its meaning depends upon the context, where does that leave the judge? Does it mean that unless I retreat to narrow legalism, a theory I reject, I have to go about my work on the basis that law is inherently indeterminate and that I can do as I please? HLA Hart aptly described these two extremes as, on the one hand, the realist ‘nightmare’ that judges never decide according to the law and, on the other hand, the idealist ‘noble dream’ that judges always decide according to the letter of the law.9

Some judges embrace as liberating the idea that the law is indeterminate. But others find the idea of indeterminacy unsettling and threatening. Probably most judges, including me, fit somewhere in between these two extremes. They reject the unthinking, narrow, and rigid formalism that prevailed during the early years of the twentieth century. At the same time, they realize that the abandonment of narrow legal formalism leaves them and the law they administer adrift and in constant search for an anchor to legitimate their task. As Michael Kirby, formerly a member of the High Court of Australia, put it, we search for a point ‘somewhere (p.96) between the spectacle of a judge pursuing political ideas of his or her own … irrespective of the letter of the law, and the unrealistic mechanic deified by the strict formalists’.10

The practical certainty of law

Now it is important to put the question of judicial decision-making into proper perspective. We should not exaggerate the significance of the law’s uncertainty by focusing exclusively on the most difficult and contentious cases that come before apex courts. As a practical matter, for most disputes, the law does provide a discernible rule by which the parties can govern their affairs. Virtually all social and economic interactions and transactions proceed without litigation on the basis of legal rules that the parties can identify, accept, and follow.11 When disputes do arise, most settle before legal proceedings are initiated and, even when the parties do find themselves in litigation, most cases settle before trial. Of those that go to trial, most involve contested issues of fact, not law. Few trial decisions are appealed. So by focusing on difficult appellate decisions, we are looking at the very tip of the pyramid of disputes, and ignoring the vast bulk of readily resolved disputes that lie at the base.

I sit in an intermediate appellate court and a very high percentage of the cases we decide, I would say 80 to 85 per cent, fall into the category of ‘error-correction’. These are cases where the issue is whether or not the trial judge erred in the application of a settled legal principle. In these cases, the law, combined with the trial judge’s findings of fact and credibility, channels the debate and our decisions have little or no jurisprudential significance.

We should not become too obsessed with hard cases, the difficult, cutting-edge cases where uncertainly reigns. Our encounters with trees—the hard cases—may have caused us to lose sight of the forest—the law.12

Hard cases—Is there a right answer?

It must be admitted, however, that there are cases where the correct legal result is anything but clear. These are the cases that catch the headlines and that perplex (p.97) legal theorists in search of a sound and coherent explanation for judicial decision-making. The prime example is decisions on fundamental rights and freedoms. What do we mean by equality? Liberty? Freedom of expression? The precise meaning of these rights and freedoms has been a matter of debate among politicians, lawyers, and moral philosophers for centuries and it is doubtful that we will ever arrive at a complete or settled meaning.

We realize that the direction the law takes in such cases is contingent on many factors: when the case comes before the court, who argues it, who decides it,13 and the political climate of the day.

It is here that the once-radical but now accepted critique of the formalist model comes to the fore. I think it is important for judges to confront this difficult issue rather than hide behind the myth of strict legalism. The role of the judge is not, as Chief Justice John Roberts infamously claimed, simply that of the baseball umpire who simply calls balls and strikes.14 Myths and fictions of that kind are dangerous, not only because they are wrong, but also because they discourage judges from being honest and reflective about the values they bring to judging.

If judges really believe that their personal views cannot influence their decisions, they will fail to make the necessary effort to put those personal biases and prejudices aside. The conscientious judge will not hide behind a façade of judicial neutrality but will engage in self-reflection, consciously striving to confront the influence of personal views and attitudes when making decisions. The judge should not pretend to be an amoral and apolitical automaton.

HLA Hart argued that in hard cases, when the law does not dictate an answer, the judge has the discretion to decide the cases ‘according to his own beliefs and values’ and ‘to follow standards or reasons for decision which are not dictated by the law’.15

I accept that when the law does not dictate a clear answer, I do have a choice. But I am unable to infer that my decision-making responsibility can be accurately described as choosing from a range of equally acceptable results on the basis of my personal views and values. Indeed, Hart himself conceived of the judge’s discretion as being circumscribed. He insisted that the choice the judge makes must not be arbitrary and that it should ‘display characteristic judicial virtues’ including ‘a concern to deploy some acceptable general principle as a reasoned basis for decision’.16

I accept that any claim that we can confidently predict or know ‘the right answer’ to every case that comes before the courts is doomed to fail. Different judges see (p.98) legal issues in different ways and different judges, acting in perfect good faith and sharing an eager desire to do justice, will disagree. But when I sit down to write my reasons, I find it difficult to describe my approach as being other than to do my best to come to the legally correct decision. It seems to me that correct results are what the legal system aspires to achieve and that my working hypothesis has to be that I am in pursuit of the right answer, even though I know that I may not be able to claim with confidence that I have found it.

Many find this answer unsatisfactory. Even Ronald Dworkin, who argued that there were right answers, was driven to rely on the Hercules construct, ‘a lawyer of superhuman skill, learning, patience and acumen’17 to decide the ‘hard cases’ that cannot be decided ‘under a clear rule of law, laid down by some institution in advance’.18 I am no Hercules and my skill, learning, patience, and acumen are far from superhuman, yet I still must decide the hard cases.

I think that it is important for judges to focus on the noble dream, the ideals and aspirations of law, not on our shortcomings and human failings. That I am no Hercules does not relieve me from thoroughly performing my Herculean task. To carry it out in a manner acceptable in a democratic society, I must strive to identify and remain faithful to a vision of law as a coherent set of norms and standards. As a judge I work within an institutional framework that aims to achieve that aspiration: a framework that guides and disciplines the process of judicial decision-making and that seeks to limit the influence of the personal views of the judge. I have to be honest and admit that sometimes the framework only takes me so far, and that I could credibly decide the case either way. I concede that the law does not dictate one right answer, but I aspire to attain a result that displays what Dworkin calls integrity, the idea that ‘the law is structured by a coherent set of principles about justice and fairness and procedural due process’ and that judges must enforce those principles in every case ‘so that each person’s situation is fair and just according to the same standard’.19

I see the decision-maker’s task as being constrained by the need to find a result that best comports with the fabric and texture of the legal rules and principles pertinent to the dispute. The integrity of the legal process compels me to define my responsibility in terms of providing a reasoned justification for my decision that aims to persuade the litigants, the legal community, and the public that I have reached the right result, the result that achieves justice under law, not justice according to my personal beliefs.

In hard cases, my decision will make new law. That is an awesome responsibility. To carry it out in a manner acceptable in a democratic society, I must strive to identify and remain faithful to a vision of law as a coherent set of norms and standards. (p.99) As a judge, I work within an institutional framework and a process that aims to achieve that aspiration, a framework that guides and disciplines the process of judicial decision-making.

Let me describe some of the constraints that guide me in this difficult task.

The ‘artificial reason’ of the law

Legal reasoning has much in common with forms of argument used outside the law.20 Lawyers and judges use the methods of philosophy, reasoning by analogy, and logical progression from moral and ethical principles. To discern the meaning of legal doctrines, they look to history and the way the law has evolved over time. And like social scientists, they examine legal doctrine from the perspective of social values, mores, and welfare.

But while lawyers and judges draw on other disciplines to do their work, legal reasoning has distinctive features.21

Philosophy and the social sciences have much to teach about general principles of justice. But lawyers and judges have the special skills required to craft workable legal standards by which specific disputes may be decided in a fair and just manner. The distinctive task of the law is to move from the general to the particular and to deal with the crucially important details about the fair resolution of disputes, the precise contours of legal rights and their enforcement at the level of particular cases. So, for example, the law’s commitment to the over-arching principle that promises should be kept derives from moral philosophy and social sciences such as economics. However, the important details of contract law are derived from on the ground experience of lawyers and judges resolving issues on a case by case basis.

Legal reasoning operates within a defined framework that requires arguments to have a pedigree that qualifies them as legitimate. This limits the bite of arguments lacking that pedigree and based on factors external to it. Fundamental to the very idea of law and to the nature of legal argument is that a judge’s choice when deciding a case is constrained. Simply put, a judge is not free to decide cases on the basis of what seems to the judge, all things considered, to be the best possible outcome. The judge is not allowed to consider all things. The judge is charged with finding the result that best fits with the relevant legal rules and principles, not the result that comports with the judge’s personal view of right and wrong.

Scientists, philosophers, and social scientists are engaged in a search for the truth unconstrained by authority. Legal reasoning is different. Sir Edward Coke, the great legal scholar and judge who fought to establish the common law constitution and (p.100) the independence of the judiciary in the early seventeenth century, spoke of the ‘artificial reason’ of the law, which he distinguished from ‘natural reason’. Refuting James I’s claim to be the font of all justice, Coke argued that even if the King was endowed with ‘excellent science’, the rights of the King’s subjects had to be decided not by the King using natural reason but by the judges using the ‘artificial reason and judgment of law’.22

The ‘artificial reason’ of the law does not put judges and lawyers in an impossible straightjacket of strict legalism. Legal standards, I have suggested, are by nature general and open-ended. They afford judges considerable latitude to reach just results in individual cases and there is certainly scope for judges to develop the law to make it more just. The wise judge will take account of the moral, historical, and philosophical roots of the legal standards the judge is interpreting and applying. The wise judge will also probe and prod to see if there are not holes to be poked in a formal legal argument that leads to an unjust result.23

But wise judges must also recognize that the artificial reason of the law imposes limits on their capacity to make the world a better place. The ‘pervasive formality of law’—and the ‘tendency to take its rules and their words seriously even though in some cases they might work and injustice—is what distinguishes law from many other decision-making contexts’.24

So while I reject narrow legalism, I accept that there is a necessary element of formalism in judicial decision-making. Legal doctrine does matter and the law limits my choices even when I am trying to achieve justice in the individual case. I have to apply constitutionally valid statutes as they are written. To take a familiar example, I may not like statutory mandatory minimum sentences that require me to impose a sentence I consider to be unduly harsh, but if the statute passes constitutional muster, I have no choice.

The doctrine of stare decisis requires me to follow binding precedents even if I do not like the result. For example, there is a common law rule requiring ‘consideration’, a term indicating the exchange of something of value, to support the enforcement of a contract. Suppose that I think the need for consideration is outmoded and that all promises should be enforced. I hear a case where enforcement of a promise is resisted on the ground that there was no consideration for the promise. If I cannot fit the facts of the case into a recognized exception to the (p.101) traditional consideration rule, I have no choice. Much as it may pain me to render what I consider to be an unjust result, I cannot enforce the contract.25

Even when deciding a hard case for which the precedents provide no clear answer, the doctrine of precedent constrains. If I make a decision in a ‘hard case’, my decision stands as a precedent that I, and all the other judges at or below my level in the judicial hierarchy, must follow in the future. The fact that my pronouncement will bind me, my colleagues, and the lower courts, makes me cautious. It significantly constrains my power to decide the case on the basis of my personal views and it reinforces my obligation to decide in a principled legal fashion, one that best fits with the established law and the prevailing norms and standards of our legal system.

Collegial decision-making

Another constraint is that as a judge of an appellate court, virtually every decision I make is reached through a collegial process. The process of collegial decision-making is a strong counter-weight to any tendency to be influenced by values extraneous to the law. It requires me to engage not only with the arguments presented by the litigants and the lawyers but also with any opposing views from my colleagues when we debate the case in the conference room and as we prepare our reasons for judgment. The collegial process ensures that the extraneous personal views of each individual judge are diluted; in my court, by two and sometimes four other judges and at the Supreme Court of Canada by eight others. Related to this is a kind of ‘team spirit’ that prevails in most collegial courts. We share a common interest in maintaining and enhancing the reputation of the institution as stable, reliable, independent, and committed to the legal integrity of its work.

Respecting the boundaries of decision-making authority

Another important constraint on judicial authority is imposed by the allocation of decision-making authority to other decision-makers. Judges operate within a legal and constitutional order that defines the decision-making powers of the various actors.

Judges must respect the decision-making authority conferred on other adjudicators. Administrative agencies and tribunals are empowered to decide myriad (p.102) issues and disputes. Provided those agencies and tribunals are acting reasonably and within the limits of the powers they enjoy, judges must accept their decisions, even in cases where they would have decided otherwise. Trial judges are charged with the duty to find the facts and appellate judges must resist the temptation to retry the case on appeal. Here we encounter the competing principles of deference and legality.

Deference defines the measure of respect the reviewing or appellate court must accord the decision under review. Deference reminds the reviewing or appellate court that first-instance decision-makers have certain institutional advantages with respect to questions of fact, policy, and certain questions of law that are central to their mandate. Deference instructs reviewing or appellate courts to refrain from approaching the case as if they were the first-instance decision-maker. First-instance decision-makers are given a margin of appreciation, protecting their decisions from unduly strict scrutiny. This means that even if the reviewing or appellate court would have decided the case differently, ordinarily, it may not interfere with the decision unless it can say that the decision was unreasonable or infected by certain kinds of legal error.

Deference also reflects legislative supremacy and the right of Parliament and the legislatures to assign decision-making powers to administrative agencies, for example, human rights commissions, labour boards, and immigration and refugee tribunals. Provided the legislation that assigns decision-making authority does not infringe upon the constitutionally protected authority of the courts,26 the legislator’s choice must be respected.

But deference has its limits and at a certain point, confronts the competing principle of legality and the ideal of a uniform standard of justice. A reviewing or appellate court must reconcile the need to defer to first-instance decisions with the duty to ensure the overall legal integrity of the decision-making process. Reviewing and appellate courts are instructed by the principle of deference to refrain from unduly meddling with specific decisions but, at the same time, are encouraged by the principle of legality to intervene when first instance decisions do not respect the law’s general standards.

Drawing the line between the principles of deference and legality has proven to be a notoriously difficult task. While this may sound like a purely technical question, it is probably the question that gnaws at my judicial conscience more than any other. The reason for that is very simple. As a judge, I am committed to the idea that every case should be decided justly. Yet the standard of review and the principle of deference instruct me on a daily basis to keep my appellate judicial hands off decisions that I would have made differently, decisions that I think are wrong but not so wrong that I can call them unreasonable.

(p.103) My primary role as an appellate judge is to define, explain, and elucidate the law and to ensure its uniform application. When filtered through the screen of deference, this means that as an appellate judge, I apply a standard of correctness to a trial judge’s conclusions in law. If I identify a legal error in the trial judge’s decision, I can correct it.

On the other hand, I must accept and respect the trial judge’s primary role of deciding specific disputes on their facts. I must defer to a trial judge’s findings of fact and I may only reverse factual findings that exhibit a demonstrable error at the heart of the case, or to use the technical label, a ‘palpable and overriding error’.27 A palpable error is one that is obvious. For example, if there is no evidence to support a factual finding, it is obviously wrong. If one finding conflicts with another finding there is an obvious error. Speculative findings and findings based on a misapprehension of the evidence are further examples of obvious or palpable errors.28 An ‘overriding’ error is one that is of sufficient gravity to ‘vitiate the challenged finding of fact’, in other words, an error that ‘goes to the root of the challenged finding of fact such that the fact cannot safely stand in the face of that error’.29 Errors of this nature implicate the integrity of the legal process and accordingly are not protected from appellate review. So if the finding is ‘contrary, not merely to what the appellate court would have done, but contrary to what any trier of fact could reasonably have done’,30 the appellate court must intervene.

There are good reasons for the division of labour between trial and appellate courts. Let me first consider questions of law. It is basic to a sound system of justice that the law be applied uniformly to all. This is sometimes referred to as ‘the principle of universality’ that ensures ‘that the same legal rules are applied in similar situations’.31 Appellate courts have the duty to monitor and supervise the legal integrity of decision-making. They have an overview of the legal system and are ordinarily better placed to decide points of law than a single trial judge who is necessarily immersed in the facts and details of the specific case. Appellate decisions have precedential effect and will be binding in the future. This means that appellate judges have to focus not only on the specifics of the case but on the broader implications of the decision. This gives rise to a certain expertise in assessing how a specific ruling in one case may affect other cases in the future. Appellate courts also sit in panels of three or more judges. This ensures that points of law are decided on a collegial basis and it tends, thereby, to insulate decisions from the personal or idiosyncratic views of one judge.

(p.104) When we turn to questions of fact, it is widely accepted that trial judges have the advantage. Trial judges are ‘in the trenches’ and on the front-lines of justice. They see and hear the parties and the witnesses. They watch the narrative of the case unfold and they guide the case from its start to its completion. The trial judge lives with the case for the duration of the trial, often days or weeks. This puts the trial judge in ‘a privileged position’32 to deal with issues of fact. By dealing with the dispute hands-on, trial judges gain an intimate understanding of the case. This first-hand perspective is impossible to replicate on appeal. In the words of an author cited with approval by the Supreme Court of Canada, the trial judge is ‘exposed to the entire case’ and gains a ‘total familiarity with the evidence’ which facilitates a ‘far deeper’ insight than is possible on appeal.33 An appellate judge can only read the paper record. Oral argument on appeal will be time-limited. The view of the case is ‘telescopic’34 or ‘myopic’35 rather than at large.

Appeal courts are required to give deference to the trial judge’s findings of fact for three reasons.36

First, it is unlikely that the quality of decisions on questions of fact would be improved by allowing open-ended appellate review. There is a significant risk that without the first-hand familiarity with the case that can only be gained by hearing and seeing the witnesses, appellate decisions on factual questions will be of a lower quality than those made by trial judges. So, however paradoxical it might seem, the curtailment of appeal rights and the restraint imposed on appellate judges from interfering with factual findings is motivated by a desire to get to the right result.

Second is the interest in efficiency and minimizing the cost of litigation. As there is nothing to be gained in quality, the interests of efficiency strongly favour limiting appeals on questions of fact. It would be costly and time-consuming to give appellate judges the time and resources they would require to be in the position where they could safely second-guess trial judges on questions of fact. And even then, it is unlikely that a full-blown appeal on factual issues would yield better or even equally sound results. Saddling the parties, especially litigants with limited resources, with the cost of such an appeal would be both inefficient and unfair.

I am generally wary of allowing systemic efficiency to trump individual justice. However, I accept that if the justice system is too costly, access to justice is lost and with lost access goes the protection of individual rights. Appeals are expensive and can be used by wealthy litigants to wear down their opponents. Modern civil procedure emphasizes the principle of proportionality, the idea that the nature and cost of the process should be proportionate to the value, complexity, and (p.105) importance of the case so that civil justice is more timely, efficient, and affordable. This curtails the unbridled pursuit of ‘justice at all costs’ that can put justice out of reach. Open-ended appeals could undermine the principle of proportionality and the encouragement of cost-effective justice.

The third reason for limiting appeals on questions of fact is promoting the integrity of the trial process. Allowing appeal courts to review all trial decisions on a correctness standard could undermine public confidence in the trial process. Trials would become nothing more than the first crack and if one party did not like the result they could have a second crack at the Court of Appeal. As the Supreme Court of Canada put it, an appeal should be ‘the exception rather than the rule’.37

Trial judges have primary responsibility to find the facts. Appellate judges must respect that authority, unless the finding was unreasonable, even when they think the trial judge got the facts wrong. Many decisions by first instance decision-makers are discretionary. A judge who sits higher in the legal hierarchy may disagree with the first instance decision-maker’s exercise of discretion, but if it falls within the margin of appreciation allowed by the law, the reviewing judge must swallow any personal misgivings and uphold the decision.

The obligation imposed on appellate courts to defer to the factual findings and discretionary decisions of trial judges represents a very significant discipline on appellate decision-making. Equally, the obligation of trial judges to base their decisions on the law as laid down by appellate courts disciplines the choices they make.

The evidence, the record, and the parameters of the dispute

Judges are also constrained by the way the parties have defined the dispute and by the evidence the parties have led. Party definition of the issues and presentation of the facts are fundamental features of the adversarial system. Judges do not define or shape the scope of the dispute they are to decide. That is left to the parties who have the right to plead their case as they see fit. The parties also control the evidence that is put before the court. It is only in rare and exceptional cases that a trial judge is permitted to insist that a witness be called. If a litigant decides not to advance an issue or an argument, judges almost always respect the litigant’s choice, even if the judge would have presented the case differently. While modern procedural rules involve more and more judicial case-management, the parties still retain essential control over the definition of the issues and decide what evidence and arguments to present.

On appeal, the trial record is all-important and significantly constrains the appellate court. Appeal judges are reluctant to consider arguments or facts not (p.106) introduced before the trial court. Allowing a losing litigant to shift ground and to advance a new argument on appeal would be unfair to the opposing party. It would also undermine the integrity of the trial process if a party was free to retry the case on a different theory or on different facts on appeal.

The discipline of reasons

There are certainly legal scholars and many lawyers and judges who doubt the efficacy of the constraints I am describing.38 Critics acknowledge that the form of legal argument and judicial reasoning reflects those constraints, but argue that the form is little more than a convenient mask for what really drives judicial decision-making, namely, the judge’s personal moral and political views.

That brings me to what I will describe as the discipline of reasons. The obligation to give reasons is more than a legal formality. Popular opinion is increasingly sceptical of those who have and exercise authority. The culture of transparency demands justification from all who wield power in modern democratic societies. The judiciary cannot escape the kind of scrutiny to which other public officials are exposed. The public wants to know what values, opinions, and attitudes judges bring to bear on the task of deciding cases.

This is closely related to the erosion of confidence in the legalistic ideal of the objectivity of law. One does not have to be a legal realist or part of the critical legal studies movement to be sceptical of Chief Justice Robert’s claim that as a judge, he simply ‘calls balls and strikes’. Because of our scepticism, we do not allow the Chief Justice to simply say ‘strike’. We require him to provide a reasoned opinion based upon recognized and accepted legal sources to explain why.

The parties and the right to reasons

When the judge gives reasons, the first and, for most judges, probably the most important audience is the parties to the case and, in particular, the losing party. The parties have a right to reasons and the judge is speaking directly to them. Judges try to demonstrate as best they can that they have heard and understood the evidence, the arguments, and the relevant law, and that they have explained why one party lost and the other party won.

Litigants are entitled to reasons as a simple matter of justice and fairness. The right to a reasoned decision may be seen as an aspect of the right to be heard. As a reputed English judge observed in this regard, the most important person in the (p.107) courtroom ‘is the litigant who is going to lose’.39 Without a reasoned decision to explain why the litigant lost the case, the litigant can have no confidence that the judge actually understood and grappled with the issues. As the English Court of Appeal stated: ‘[J]ustice will not be done if it is not apparent to the parties why one has won and the other has lost.’40 Lord Denning explained that reasons give ‘proof’ that the judge ‘has heard and considered the evidence and arguments’ and ‘has not taken extraneous considerations into account’.41

The Supreme Court of Canada has located the right to reasons in ‘the dignity interest … an interest at the heart of post-World War II jurisprudence’.42 An academic commentator puts it this way: ‘[R]easons may be regarded as an integral part of treating a disappointed litigant with the respect which his dignity as a citizen demands.’43 In a like vein, another author states, that reasons have a ‘humanizing’ effect, demonstrating respect on the part of the judge towards the losing party.44 And as Sir Robert Megarry put it years earlier: ‘[T]o be condemned without being understood is as bad as to be condemned unheard.’45 We will never make the losing parties happy but a reasoned decision will be more acceptable because it demonstrates to losing parties that they were heard and that their arguments were understood and grappled with.

The appellate court and the facilitation of appellate review

For all but the judges of apex courts, the second audience when writing reasons is the Court of Appeal or the Supreme Court. Judges know that their reasons are subject to scrutiny by a higher court and that their decisions may be reversed if they reveal legal error. Judges must do their best to explain their decisions in a way that shows the higher court precisely why they decided as they did so that the higher court can exercise its oversight role and ensure that the decision comports with the law.

The requirement for reasons is formally cast by appeal courts as a means to facilitate appellate review.46 From that perspective, reasons have an informational effect. They let the parties know why the decision was made so that they can decide whether to take the next step and appeal to a higher court. Reasons also provide (p.108) appellate courts with the tools they need to review trial decisions. An appellate court needs to be able to see the basis for the trial judge’s factual findings and the train of legal reasoning that led to the decision. As the Supreme Court of Canada has explained, adequate reasons are essential to maintaining the boundary between the trial and appellate functions. The trial judge’s ‘clear articulation of the factual findings’ allows the appellate court to correct errors while at the same time ‘inhibiting appeal courts from making factual determinations “from the lifeless transcript of evidence, with the increased risk of factual error” ’.47

The public and the integrity of the legal process

The third audience, especially important for apex courts, is the public, not just the wider legal community but also the public at large. Judges have to realize that their decision-making authority is an exercise of public power and that the integrity and authority of the court on which the judge sits will suffer if its decisions are not based on legally accepted standards.

Reasons make judges publicly accountable for their decisions. The public, as well as the litigants, is entitled to an explanation. The requirement of reasons may be seen as an element of the open court principle, the idea that the public has a right to attend court to see justice openly administered. Justice must not only be done, but also seen to be done.48 As a Canadian scholar has observed, ‘reasons provide a basis for … public and political debate about the justice of the particular decision and of the underlying law that authorized it’.49

Reasons foster public confidence in the judicial process. The obligation to provide reasoned decisions forces judges to be transparent. The ‘legal culture of justification’50 comports with the more general culture of justification that pervades modern democratic life.51 Reasons open the window to why a decision was made and that, in turn, assures the public that the decision was not arbitrary. The Supreme Court of Canada sees the need to provide reasoned judgments as being ‘central to the legitimacy of judicial institutions in the eyes of the public’.52 Reasons for judicial decisions are necessary to bridge the gap between judicial power and democratic legitimacy: reasons allow unelected judges to demonstrate to the public that the judicial process is legitimate and based on sound principle and (p.109) rationality. By providing reasons and exposing to public view the basis for judicial decisions, judges provide the means for public scrutiny, accountability, and, where appropriate, criticism.

I believe that the obligation to expose one’s reasons for decision to the legal community and the public at large represents a significant constraint on the exercise of judicial power. Judges are members of the legal community and the legal culture that publicly proclaims its sense of integrity, propriety, and commitment to the rule of law. Reasons cement the judge’s adherence to those values and the obligation to abide by the limits they impose. By providing reasoned decisions, judges demonstrate that they are not, as the realists and judicial cynics claim, simply deciding on the basis of their own personal views but rather operating within the limits imposed by the legal regime and the judicial function. American Judge Patricia Wald explains that any temptation to reason backwards from desired result to rationale is inhibited on an appellate court: ‘there are judges of other persuasions to brake the momentum; there are precedents galore that must be acknowledged and accommodated; there is always the judge’s own sense of integrity toward the development of coherent law’.53 And as EW Thomas puts it, judges ‘are constrained by their adherence to the rule of law … [and] they will consciously strive to reach a decision which is acceptable within the bounds of that concept rather than self-indulgently impose their own will or ideal’.54

Reasons demonstrate that the decision was not arbitrary or based on personal whim. Reasons reflect our obligation to expose our thinking to the parties, the appeal process, the academics, the media, the public, and the politicians. If our reasons are not based upon sources and arguments accepted by the legal community or fail to abide by the relevant accepted legal norms, they will not withstand scrutiny, whether in the Court of Appeal or in the court of public opinion. Reasons thereby constrain judicial power. They limit and legitimate judicial creativity by demonstrating that judges are not simply deciding the case on the basis of their own personal views or predilections, but rather operating within the circumscribed boundaries implicit in the judicial function and imposed by the law.

Ronald Dworkin links the obligation to provide reasons with the virtue of integrity that requires all who exercise power ‘to act in a principled and coherent manner toward all citizens’.55 Reasons encourage judges ‘to conceive of the body of law they administer as a whole rather than as a set of discrete decisions that they are free to make or amend one by one’,56 to avoid arbitrary distinctions between similar cases and to base the decision on a principle that forms part of ‘a single, coherent scheme of justice and fairness’.

(p.110) So as a judge, I see my task as being constrained by the need to find a result that best comports with the fabric and texture of the legal rules and principles pertinent to the dispute. I have to accept that there are too many gaps and holes in the fabric to provide me with a ready answer. But I do my best to complete the weave in a proper and coherent manner. While I make no philosophical claim that I have found the truth, the integrity of the legal process compels me to define my responsibility in terms of providing a reasoned justification for my decision that aims to persuade the litigants, the legal community, and the public that I did my best to reach the right result.

The logic of discovery and the logic of justification

Retired South African Constitutional Court judge Albie Sachs, by no means an unduly formalistic judge, provides a compelling explanation of the discipline of reasons. Sachs once stated: ‘Every judgment I write is a lie’.57 Why a lie? Because, according to Sachs, contrary to their appearance, ‘[l]egal judgments … [do] not emerge from the dispassionate placing of logical propositions in rationally ordained sequence’ but rather from ‘an inchoate—even chaotic—mental firmament’58 with ‘an enormous amount of random intuitive searching and a surging element of unruly, free-flowing sensibility’.59 The final form of the judgment fails to reveal the complex mental process that the judge follows to come to a decision.

If Sachs stopped there judicial decision-making would be reduced to nothing more than a shallow charade and subterfuge for judges to decide cases on the basis of their own private and personal predilections.

But Sachs goes on to provide a valuable insight that significantly qualifies his claim that every judgment is a lie. Sachs identifies two phases of the judicial decision-making process. The first phase, the one that corresponds with intuition or the ‘judicial hunch’ account of decision-making, he describes as the ‘logic of discovery’. The logic of discovery describes how a judge starts the mental process of sifting through the evidence and the legal arguments to come to an initial or tentative conclusion. The second phase, the ‘logic of justification’ reveals that the product of the initial journey of discovery will not suffice—‘a discovery that cannot be justified simply cannot stand’.60

The sound and conscientious judge will not decide a case solely on the basis of an initial hunch or initial discovery of what seems to be the right result. The sound and conscientious judge must justify the result on the basis of what Sachs describes (p.111) as ‘accepted principles, rules, and standards to arrive at a conclusion that is consistent with those rules, principles and standards’.61

Every judge has had a ‘it just won’t write’ moment. After hearing the argument, at the ‘discovery’ stage, the decision seems clear. But after sitting down and trying to write reasons for a decision it becomes clear that the discovery cannot be justified. A discovery that ‘won’t write’ cannot survive as a judgment. The judicial reasoning process that starts with intuition and flashes of inspiration must end with a reasoned judgment capable of justifying the result, even if that means, in Sachs words, ‘causing [the judge] to abandon even the strongest initial intuitions …’.62 Justice Ruth Bader Ginsberg of the United States Supreme Court makes a similar observation: ‘A court charged with defining the law may not rely on unarticulated intuition … the court is obliged conscientiously to reason why’.63 Even Jerome Frank, who adhered to the ‘hunch’ school of jurisprudence, accepted that the discipline of reasons had to prevail: ‘Often a strong impression that, on the basis of the evidence, the facts are thus-and-so gives way when it comes to expressing that impression on paper.’64


So, at the end of the day, after I have rendered my judgment in a hard case and I ask myself ‘did I reach the right answer?’ I am never quite sure. But I do want to be absolutely certain that my reasons demonstrate to the litigants, to the Supreme Court, and to the public at large that the parties were fairly heard; that I grappled with the arguments they made; and that I assessed those arguments and analysed them in a manner that respects the need for legal coherence and integrity, and that comports with the accepted principles and norms of our legal culture. (p.112)


(*) An earlier version of this chapter was delivered in April 2016 at a conference on judicial decision-making at The Foundation for Law, Justice, and Society, Wolfson College, Oxford. The chapter is drawn from my book, Good Judgment: Making Judicial Decisions (University of Toronto Press 2018) with the permission of Toronto University Press.

(1) BN Cardozo, The Nature of the Judicial Process (Yale University Press 1921) 166.

(2) ibid 143.

(3) JAK Thomson (trs), Nichomachean Ethics (Penguin 1977) 1137 a-b, quoted in F Schauer, Thinking Like a Lawyer (Harvard University Press 2009) 28.

(4) ibid.

(5) See eg, K Llewellyn, The Bramble Bush (Oceana 1930); J Frank, ‘Are Judges Human?: Parts I & II’ (1931) 80 U Penn L Rev 17, 233; U Moore and T Hope, ‘An Institutional Approach to the Law of Commercial Banking’ (1929) 38 Yale LJ 703.

(6) See eg, D Kennedy, ‘Form and Substance in Private Law Adjudication’ (1976) 89 Harv L Rev 1685; R Mangabeira Unger, The Critical Legal Studies Movement (Harvard University Press 1983).

(7) See eg, C MacKinnon, Sexual Harassment of Working Women (Yale University Press 1979); J Nedelsky, ‘Reconceiving Autonomy’ (1989) 1 Yale JL & Feminism 7; ML Minow, Making All the Difference: Inclusion, Exclusion & American Law (Cornell University Press 1990); R Delgado, Critical Race Theory: An Introduction (New York University Press 2011).

(8) See eg, R Coase, ‘The Problem of Social Cost’ (1960) 3 JL & Econ 1; G Calabresi, The Costs of Accidents: A Legal and Economic Analysis (Yale University Press 1970); RA Posner, Economic Analysis of Law (Little, Brown & Company 1973); RH Mnookin and L Kornhauser, ‘Bargaining in the Shadow of the Law: The Case of Divorce’ (1979) 88 Yale LJ 950.

(9) HLA Hart, ‘American Jurisprudence through English Eyes: The Nightmare and the Noble Dream’ in Essays in Jurisprudence and Philosophy (OUP 1983) 128.

(10) MD Kirby, ‘ “Judicial Activism”? A Riposte to the Counter Reformation’ (2005) 11 Otago L Rev 1, 27.

(11) There are, however, undoubtedly some cases that settle because the law is uncertain and the parties decide to avoid the risk of litigation.

(12) See G Priest and W Klein, ‘The Selection of Disputes for Litigation’ (1984) 13 J Legal Stud 1, pointing out that as only contentious cases get litigated, they represent a biased sample of all legal events. See also, Schauer (n 3) 22.

(13) See White v Chief Constable [1998] 2 WWR 1509 at 1549 (Lord Hoffmann), describing an earlier decision as ‘one of those cases in which one feels that a slight change in the composition of the Appellate Committee would have set the law on a different course’.

(14) J Roberts, Confirmation Hearing on the Nomination of John G. Roberts Jr. to be Chief Justice of the United States, 12 September 2005 (Serial No J-109-37), Washington: US Government Printing Office, 2005.

(15) HLA Hart, The Concept of Law (3rd edn, OUP 2012) 273.

(16) ibid 201–05.

(17) R Dworkin, Taking Rights Seriously (Duckworth 1977) 105.

(18) ibid 81.

(19) R Dworkin, Law’s Empire (Harvard University Press 1986) 243.

(20) The classic discussion is B Cardozo, The Nature of the Judicial Process (Yale University Press 1921).

(21) For an excellent discussion that has helped me shape my thinking on these issues, see Schauer (n 3).

(22) Prohibitions del Roy (1607) 12 Co Rep 63, quoted by J Bickenbach, ‘The “Artificial Reason” of the Law’ (1990) 12 Inf Log 23. See also C Fried, ‘The Artificial Reason of the Law or: What Lawyers Should Know’ (1981) 60 Texas L Rev 35.

(23) See A Barak, The Judge in a Democracy (Princeton University Press 2006) 59, 67: ‘judges must not impose their own personal, subjective perceptions’ but ‘must aspire to reach just solutions.’ And if the law seems to point to a result that contradicts the judge’s sense of justice, ‘the judge must retrace his footsteps’ to see ‘whether he has strayed from the path, for law’s aspiration is to be just, and the judge’s aspiration is to do justice’.

(24) Schauer (n 3) 33.

(25) Southern Pacific Company v Jensen (1917) 244 US 205 at 221 (Holmes J): ‘A common-law judge could not say, “I think the doctrine of consideration a bit of historical nonsense and shall not enforce it in my court”.’

(26) Constitution Act, 1867, ss 96–100; Crevier v Quebec (AG) [1981] 2 SCR 220.

(27) Housen v Nikolaisen 2002 SCC 33, [2002] 2 SCR 235.

(28) Waxman v Waxman 2004 OJ No 1765 (CA) at para 296.

(29) ibid at para 297.

(30) RJ Traynor, ‘Some Open Questions on the Work of State Appellate Courts’ (1957) 24 UCLR 211, 222.

(31) Housen (n 27) para 9.

(32) Schwartz v Canada [1996] 1 SCR 254 at para 32.

(33) RD Gibbens, ‘Appellate Review of Findings of Fact’ (1991–92) 13 Adv Q 445, 446, cited in Housen (n 27) para 14.

(34) Gibbens (n 33).

(35) Traynor (n 30) 221.

(36) Housen (n 27) paras 16–18, where these reasons are identified but listed in a different order.

(37) ibid para 17.

(38) For a recent judicial perspective, see RA Posner, Divergent Paths: The Academy and the Judiciary (Harvard University Press 2016).

(39) RE Megarry, ‘Temptations of the Bench’ (1978) 16 Alta L Rev 406.

(40) English v Emery Reimbold & Strick Ltd [2002] EWCA Civ 605, [16].

(41) A Denning, The Road to Justice (Stevens and Sons 1955) 29.

(42) R v M (RE) [2008] 3 SCR 3, para 11.

(43) T Allan, ‘Procedural Fairness and the Duty of Respect’ (1998) 18 OJLS 497, 499; M Elliot, ‘Has the Common Law Duty to Give Reasons Come of Age Yet?’ [2011] PL 56, 62–63.

(44) M Cohen, ‘Reason-Giving in Court Practice: Decision-Makers’ (2007–2008) 14 Colum J Eur L 257, 258, 265.

(45) R Megarry, Lawyer and Litigant in England (Stevens and Sons 1962).

(46) R v Sheppard [2002] 1 SCR 869, 2002 SCC 26.

(47) R v REM [2008] 3 SCR 3, 2008 SCC 51 [11] quoting M Taggart, ‘Should Canadian Judges be Legally Required to Give Reasoned Decisions in Civil Cases’ (1983) 33 UTLJ 1, 7.

(48) R v Sussex Justices, Ex parte McCarthy [1924] 1 KB 256.

(49) H Stewart, ‘The Trial Judge’s Duty to Give Reasons for Judgment in Criminal Cases’ (2009) 14 Can Crim Law Rev 19, 23.

(50) D Dyzenhaus, ‘The Politics of Deference: Judicial Review and Democracy’ in M Taggart (ed), The Province of Administrative Law (Hart Publishing 1997).

(51) J Rawls, A Theory of Justice (Harvard University Press 1971) 133.

(52) R v Sheppard (n 46) para 5.

(53) P Wald ‘The Rhetoric of Results and the Results of Rhetoric’ (1995) 62 UCLR 1371, 1419.

(54) EW Thomas, The Judicial Process: Realism, Pragmatism, Practical Reasoning and Principles (CUP 2005) 248.

(55) Law’s Empire (n 19) 165.

(56) ibid 167.

(57) A Sachs, The Strange Alchemy of Life and Law (OUP 2009) 47.

(58) ibid.

(59) ibid 60.

(60) ibid 53.

(61) ibid.

(62) ibid.

(63) RB Ginsburg, ‘The Obligation to Reason Why’ (1985) 37 U Fla L Rev 205, 207.

(64) United States v Forness (1942) 125 F 2d 928 (2nd Cir) at 942.